Search

Annotating datasets for fully supervised instance segmentation tasks can be arduous and time-consuming, requiring a significant effort and cost investment. Producing bounding box annotations instead constitutes a significant reduction in this investment, but bounding box annotated data alone are not suitable for instance segmentation. This work utilizes ground truth bounding boxes to define coarsely annotated polygon masks, which we refer to as weak annotations, on which the models are pre-trained. We investigate the effect of pretraining on data with weak annotations and further fine-tuning on data with strong annotations, that is, finely annotated polygon masks for instance segmentation. The COCO 2017 detection dataset along with 3 model architectures, SOLOv2, Mask-RCNN, and Mask2former, were used to conduct experiments investigating the effect of pretraining on weak annotations. The Cityscapes and Pascal VOC 2012 datasets were used to validate this approach. The empirical results suggest two key outcomes from this investigation. Firstly, a sequential approach to annotating large-scale instance segmentation datasets would be beneficial, enabling higher-performance models in faster timeframes. This is accomplished by first labeling bounding boxes on your data followed by polygon masks. Secondly, it is possible to leverage object detection datasets for pretraining instance segmentation models while maintaining competitive results in the downstream task. This is reflected with 97.5%, 100.4% & 101.3% of the fully supervised performance being achieved with just 1%, 5% & 10% of the instance segmentation annotations of the COCO training dataset being utilized for the best performing model, Mask2former with a Swin-L backbone.

Summary: Background: Subthreshold depression is a risk factor for major depression and is associated with increased morbidity and mortality, especially in older adults. There is emerging evidence that digital interventions, including self-help interventions, may reduce depressive symptoms. We aimed to evaluate the effectiveness of a mobile messaging intervention at reducing subthreshold depressive symptoms among older adults in Brazil. Methods: PRODIGITAL was a single blind, two-arm, individually randomised controlled trial conducted in 46 primary care clinics in the city of Guarulhos, Brazil. Individuals aged 60+ years were contacted by phone following a randomly ordered list for a screening assessment. Those who presented with anhedonia and/or depressed mood (Patient Health Questionnaire (PHQ)-2≥1), and who subsequently scored between 5 and 9 on the PHQ-9 were invited to participate. The intervention arm received the ‘Viva Vida’ digital self-help intervention consisting of automated multi-media messages sent via WhatsApp. Forty-eight audio and visual messages based on psychoeducation and behavioural activation were automatically delivered over six weeks. The control arm received a single message containing information about depression. The primary outcome was the difference in mean PHQ-9 scores between treatment arms at the three-month follow-up. All primary analyses were performed according to allocated arm with imputed data. The trial is registered with ReBEC, RBR-6c7ghfd. Findings: Participants were recruited between 8 September 2021 and 19 August 2022. Of the 454 participants enrolled, 223 were randomised to the intervention arm, 231 to the control arm. Participants’ mean age was 65.3 years (SD 5.0) and 64.0% (n = 292) were female. A total of 385 (84.8%) completed the three-month follow-up assessment; no difference in mean PHQ-9 scores between the treatment arms was observed (adjusted difference: −0.61; 95% CI: −1.75, 0.53; p = 0.29). Interpretation: These results demonstrate that the Viva Vida digital self-help intervention did not help to improve subthreshold depressive symptoms amongst older adults. Further research is needed to understand why this self-help intervention was not effective in this population, and to explore how it might be adapted to achieve this goal. Funding: São Paulo Research Foundation and UK Joint Global Health Trials.

ContextThis study proposes a Bayesian network model to aid mental health specialists making data-driven decisions on suitable treatments. The aim is to create a probabilistic machine learning model to assist psychologists in selecting the most suitable treatment for individuals for four potential mental disorders: Depression, Panic Disorder, Social Phobia, or Specific Phobia.MethodsThis study utilized a dataset from 1,094 individuals in Denmark containing socio-demographic details and mental health information. A Bayesian network was initially employed in a purely data-driven approach and was later refined with expert knowledge, referred to as a hybrid model. The model outputted probabilities for each disorder, with the highest probability indicating the most suitable disorder for treatment.ResultsBy incorporating expert knowledge, the model demonstrated enhanced performance compared to a strictly data-driven approach. Specifically, it achieved an AUC score of 0.85 vs 0.80 on the test data. Furthermore, we evaluated some cases where the predictions of the model did not match the actual treatment. The symptom questionnaires indicated that these participants likely had comorbid disorders, with the actual treatment being proposed by the model with the second highest probability.ConclusionsIn 90.1% of cases, the hybrid model ranked the actual disorder treated as either the highest (67.3%) or second-highest (22.8%) on the test data. This emphasizes that instead of suggesting a single disorder to be treated, the model can offer the probabilities for multiple disorders. This allows individuals seeking treatment or their therapists to incorporate this information as an additional data-driven factor when collectively deciding on which treatment to prioritize.

BackgroundIdentifying individuals with depressive symptomatology (DS) promptly and effectively is of paramount importance for providing timely treatment. Machine learning models have shown promise in this area; however, studies often fall short in demonstrating the practical benefits of using these models and fail to provide tangible real-world applications. ObjectiveThis study aims to establish a novel methodology for identifying individuals likely to exhibit DS, identify the most influential features in a more explainable way via probabilistic measures, and propose tools that can be used in real-world applications. MethodsThe study used 3 data sets: PROACTIVE, the Brazilian National Health Survey (Pesquisa Nacional de Saúde [PNS]) 2013, and PNS 2019, comprising sociodemographic and health-related features. A Bayesian network was used for feature selection. Selected features were then used to train machine learning models to predict DS, operationalized as a score of ≥10 on the 9-item Patient Health Questionnaire. The study also analyzed the impact of varying sensitivity rates on the reduction of screening interviews compared to a random approach. ResultsThe methodology allows the users to make an informed trade-off among sensitivity, specificity, and a reduction in the number of interviews. At the thresholds of 0.444, 0.412, and 0.472, determined by maximizing the Youden index, the models achieved sensitivities of 0.717, 0.741, and 0.718, and specificities of 0.644, 0.737, and 0.766 for PROACTIVE, PNS 2013, and PNS 2019, respectively. The area under the receiver operating characteristic curve was 0.736, 0.801, and 0.809 for these 3 data sets, respectively. For the PROACTIVE data set, the most influential features identified were postural balance, shortness of breath, and how old people feel they are. In the PNS 2013 data set, the features were the ability to do usual activities, chest pain, sleep problems, and chronic back problems. The PNS 2019 data set shared 3 of the most influential features with the PNS 2013 data set. However, the difference was the replacement of chronic back problems with verbal abuse. It is important to note that the features contained in the PNS data sets differ from those found in the PROACTIVE data set. An empirical analysis demonstrated that using the proposed model led to a potential reduction in screening interviews of up to 52% while maintaining a sensitivity of 0.80. ConclusionsThis study developed a novel methodology for identifying individuals with DS, demonstrating the utility of using Bayesian networks to identify the most significant features. Moreover, this approach has the potential to substantially reduce the number of screening interviews while maintaining high sensitivity, thereby facilitating improved early identification and intervention strategies for individuals experiencing DS.

Supervised deep learning methods have produced state-of-the-art results with large labeled datasets. However, accessing large labeled datasets is difficult in medical image analysis because of a shortage of medical experts, expensive annotations, and privacy constraints in the healthcare domain. Self-supervised learning is a branch of machine learning that exploits unlabeled data to encourage network weights toward a valid latent representation of the data during a so-called pretext task. The features learned by the model while solving pretext tasks are transferred to a downstream task where limited annotations are available. In this work, we propose PatchLoc, a novel pretext task whose objective is to find the location of a given patch from an image as a source of supervision. We validated the effectiveness of PatchLoc on a downstream segmentation task using three different medical datasets. PatchLoc yields substantial improvements compared to U-Net trained from scratch and other pretext task-based approaches in a low data regime.

Graph Neural Networks (GNNs) offer a promising direction for medical image analysis, particularly due to their ability to capture complex relationships and handle non-Euclidean data structures often encountered in this domain. This study aims to demonstrate the potential of GNNs in medical image classification, showcasing their comparable performance to leading CNNs while using significantly fewer parameters. We introduce a novel GNN model, Graph Convolution Neural Network Enhanced Connectivity (GCNN-EC), incorporating an enhanced connectivity technique that leverages interrelationships between RGB channel features. Evaluated on the MedMNIST dataset, a standardized benchmark for medical image analysis, GCNN-EC performs competitively against pre-trained DNNs, highlighting its potential as an efficient and effective approach for medical image analysis. Importantly, we do not claim that GCNN-EC outperforms all CNNs in every scenario, but rather aim to showcase the potential of GNNs as a valuable tool in the medical imaging toolkit that can be used on its own or in conjuction with CNNs to further advance the state of the art. Additionally, our model outperforms traditional GNNs on standard image datasets like MNIST and CIFAR-10, mitigating the common issue of over-smoothing. These findings underscore the effectiveness of GCNs with dynamic filtering and encourage further research into GNN-based approaches for medical imaging and broader computer vision tasks.

As the global production of waste continues to rise, there is a growing demand for more effective waste management strategies to handle this expanding problem. Recycling rates in the United States for recyclable materials are below 35%, resulting in elevated levels of waste being sent to landfills. This situation has alarming consequences, contributing to rising pollution in both soil and aquatic ecosystems, and is a significant source of concern for environmental scientists and the general population alike. The presence of contamination in recycling collection trucks is a root cause impacting recycling rates, leading to the rejection of entire loads from recycling processing sites. This problem can be alleviated by automatically detecting contamination in recyclable waste that is loaded into the collection truck hopper before compaction. In this paper, we have used different state-of-the-art computer vision-based models such as Faster-RCNN, Cascade-RCNN, Retinanet, YOLOv8 and Mask-RCNN to identify contamination within a densely cluttered waste environment. We further investigate the viability of transfer learning, comparing it to the models trained from scratch. The YOLOv8-x model attained a mean-average precision of 0.395 without using transfer learning, whereas with the incorporation of transfer learning, its performance increased to 0.463.

Deep learning-based computer vision models are typically data-hungry, resulting in the rise of dataset sizes. The consensus for computer vision datasets is that larger datasets lead to better model performance. However, the quality of the datasets is often not considered. Annotating datasets for fully supervised object detection and instance segmentation tasks requires a significant investment in time, effort, and cost. In practice, due to the large sample sizes needed, this often leads to inaccuracies in the annotation process. This research aims to understand and quantify the impact of annotation quality and quantity on the performance of object detection and instance segmentation models. Specifically, the research aims to investigate how introducing additional data with varying levels of annotation quality affects mean average precision (mAP) performance. To investigate the relationship between annotation quality and quantity, subsets of the COCO and ADE20K datasets are used. For each of the datasets, three types of annotation uncertainty are added to the annotations, which are localization uncertainty, incorrect class labels, and missing annotations. Mask-RCNN, YOLACT, and Mask2Former models are trained on a variety of sample sizes for varying levels of annotation uncertainties. The results indicate there is utility in adding additional data of lesser annotation quality. The extent of the benefits of the additional data is directly related to how degraded the annotations’ are. Furthermore, the results show that all three annotation uncertainties negatively affect mAP performance, with incorrect class labels degrading mAP performance the most, followed by missing annotations and lastly localization uncertainty.

Anxiety in older individuals is understudied despite its prevalence. Investigating its occurrence can be challenging, yet understanding the factors influencing its recurrence is important. Gaining insights into these factors through an explainable, probabilistic approach can enhance improved management. A Bayesian network (BN) is well-suited for this purpose. This study aimed to model the recurrence of anxiety symptomatology in an older population within a five-month timeframe. Data included baseline socio-demographic and general health information for older adults aged 60 years or older with at least mild depressive symptoms. A BN model explored the relationship between baseline data and recurrent anxiety symptomatology. Model evaluation employed the Area Under the Receiver Operating Characteristic Curve (AUC). The BN model was also compared to four machine learning models. The model achieved an AUC of 0.821 on the test data, using a threshold of 0.367. The model demonstrated generalisation abilities while being less complex and more explainable than other machine learning models. Key factors associated with recurrence of anxiety symptomatology were: “Not being able to stop or control worrying”; “Becoming easily annoyed or irritable”; “Trouble relaxing”; and “depressive symptomatology severity”. These findings indicate a prioritised sequence of predictors to identify individuals most likely to experience recurrent anxiety symptomatology.

Fully-supervised object detection and instance segmentation models have accomplished notable results on large-scale computer vision benchmark datasets. However, fully-supervised machine learning algorithms’ performances are immensely dependent on the quality of the training data. Preparing computer vision datasets for object detection and instance segmentation is a labor-intensive task requiring each instance in an image to be annotated. In practice, this often results in the quality of bounding box and polygon mask annotations being suboptimal. This paper quantifies empirically the ground truth annotation quality and COCO’s mean average precision (mAP) performance by introducing two separate noise measures, uniform and radial, into the ground truth bounding box and polygon mask annotations for the COCO and Cityscapes datasets. Mask-RCNN models are trained on various levels of noise measures to investigate the performance of each level of noise. The results showed degradation of mAP as the level of both noise measures increased. For object detection and instance segmentation respectively, using the highest level of noise measure resulted in a mAP degradation of 0.185 & 0.208 for uniform noise with reductions of 0.118 & 0.064 for radial noise on the COCO dataset. As for the Cityscapes datasets, reductions of mAP performance of 0.147 & 0.142 for uniform noise and 0.101 & 0.033 for radial noise were recorded. Furthermore, a decrease in average precision is seen across all classes, with the exception of the class motorcycle. The reductions between classes vary, indicating the effects of annotation uncertainty are class-dependent.

BackgroundAnhedonia and depressed mood are considered the cardinal symptoms of major depressive disorder. These are the first 2 items of the Patient Health Questionnaire (PHQ)–9 and comprise the ultrabrief PHQ-2 used for prescreening depressive symptomatology. The prescreening performance of alternative PHQ-9 item pairings is rarely compared with that of the PHQ-2. ObjectiveThis study aims to use machine learning (ML) with the PHQ-9 items to identify and validate the most predictive 2-item depressive symptomatology ultrabrief questionnaire and to test the generalizability of the best pairings found on the primary data set, with 6 external data sets from different populations to validate their use as prescreening instruments. MethodsAll 36 possible PHQ-9 item pairings (each yielding scores of 0-6) were investigated using ML-based methods with logistic regression models. Their performances were evaluated based on the classification of depressive symptomatology, defined as PHQ-9 scores ≥10. This gave each pairing an equal opportunity and avoided any bias in item pairing selection. ResultsThe ML-based PHQ-9 items 2 and 4 (phq2&4), the depressed mood and low-energy item pairing, and PHQ-9 items 2 and 8 (phq2&8), the depressed mood and psychomotor retardation or agitation item pairing, were found to be the best on the primary data set training split. They generalized well on the primary data set test split with area under the curves (AUCs) of 0.954 and 0.946, respectively, compared with an AUC of 0.942 for the PHQ-2. The phq2&4 had a higher AUC than the PHQ-2 on all 6 external data sets, and the phq2&8 had a higher AUC than the PHQ-2 on 3 data sets. The phq2&4 had the highest Youden index (an unweighted average of sensitivity and specificity) on 2 external data sets, and the phq2&8 had the highest Youden index on another 2. The PHQ-2≥2 cutoff also had the highest Youden index on 2 external data sets, joint highest with the phq2&4 on 1, but its performance fluctuated the most. The PHQ-2≥3 cutoff had the highest Youden index on 1 external data set. The sensitivity and specificity achieved by the phq2&4 and phq2&8 were more evenly balanced than the PHQ-2≥2 and ≥3 cutoffs. ConclusionsThe PHQ-2 did not prove to be a more effective prescreening instrument when compared with other PHQ-9 item pairings. Evaluating all item pairings showed that, compared with alternative partner items, the anhedonia item underperformed alongside the depressed mood item. This suggests that the inclusion of anhedonia as a core symptom of depression and its presence in ultrabrief questionnaires may be incompatible with the empirical evidence. The use of the PHQ-2 to prescreen for depressive symptomatology could result in a greater number of misclassifications than alternative item pairings.

Abstract Background Depression in older adults is a challenge for health systems in most low- and middle-income countries (LMICs). Digital strategies for the management of this condition have been emerging worldwide, but the effectiveness of most of them is still unclear, especially among older adults. Thus, we aim to assess the effectiveness and cost-effectiveness of a digital psychosocial intervention to treat depression among older adults living in socioeconomically deprived areas in Guarulhos, Brazil. Methods We will conduct a two-arm individually randomised controlled trial with 1:1 allocation ratio. Five hundred older adults aged 60 years or over with depressive symptomatology (9-item Patient Health Questionnaire score, PHQ-9 ≥ 10) and registered with one of the primary care clinics will be recruited to participate in this study. A 6-week digital psychosocial programme, named Viva Vida, will be delivered via WhatsApp to participants allocated to the intervention arm. The Viva Vida will send psychoeducational and behavioural activation audio and visual messages 4 days a week for 6 weeks. The control arm will only receive a single message with general information about depression. The primary outcome will be the proportion of depression recovery (PHQ-9 < 10) assessed at 3 months. The cost-effectiveness of the intervention will be assessed at 5 months. A detailed process evaluation will be used to explore context and important implementation outcomes. Discussion This programme was based on the PROACTIVE intervention and designed to be delivered without face-to-face contact. If effective, it could be a simple treatment option, appropriate not only when social distancing is required, but it could also be included as a regular public health programme to initiate depression treatment, particularly in LMICs where resources allocated to mental health are scarce. Trial registration Registro Brasileiro de Ensaios Clínicos (ReBEC), RBR-4c94dtn. Registered on 22 October 2021 (submitted on 03 August 2021).

As the amount of waste being produced globally is increasing, there is a need for more efficient waste management solutions to accommodate this expansion. The first step in waste management is the collection of bins or containers. Each bin truck in a fleet is assigned a collection route. As the bin trucks have a finite amount of storage for waste, accepting overfilled bins may result in filling this storage before the end of the collection route. This creates inefficiencies as a second bin truck is needed to finish the collection route if the original becomes full. Currently, the recording and tracking of overfilled bins is a manual process, requiring the bin truck operator to undertake this task, resulting in longer collection route durations. To create a more efficient and automated process, computer vision methods are considered for the task of detecting the bin status. Video footage from a commercial collection route for two bin types, automated side loader (ASL) and front-end loader (FEL), was utilized to create appropriate computer vision datasets for the task of fully supervised object detection and instance segmentation. Selected state-of-the-art object detection and instance segmentation algorithms were used to investigate their performances on this proprietary dataset. A mean average precision (mAP) score of 0.8 or greater was achieved with each model, reflecting the effectiveness of using computer vision as a tool to automate the process of recording overfilled bins.

Abstract Background Depression is a common condition in older adults, being often detected and treated initially in primary care. Collaborative care models including, for example, task-shifting and stepped-care approaches have been investigated to overcome the current scarcity of strategies and trained mental health professionals to treat depression. The PROACTIVE study developed a psychosocial intervention, which makes extensive use of technology in an intervention delivered mainly by non-specialists to treat older adults with depression. The aim of this qualitative study is to assess: 1. Health workers’ fidelity to the intervention protocol; 2. Acceptability of the psychosocial intervention from the viewpoint of older adult participants; and 3. Perceptions of the psychosocial intervention by the health workers. Methods Qualitative methods were used to achieve our aims. The sample included participants (N = 31) receiving the intervention in the pilot trial and health workers (N = 11) working in a Basic Health Unit in the northern area of São Paulo, Brazil. Focus group, non-participant observation and structured interviews were used. Data were analysed using a thematic analysis approach. Results 1. Health workers’ fidelity to the intervention protocol: training, supervision and the structured intervention were crucial and guaranteed health workers’ fidelity to the protocol. 2. Acceptability of the psychosocial intervention from the viewpoint of older adult participants: Collaborative care, task-shifting, and stepped-care approaches were well accepted. The structured protocol of the intervention including different activities and videos was important to adherence of older adult participants 3. Perceptions of the psychosocial intervention by the health workers: It was feasible to have the home psychosocial sessions conducted by health workers, who are non-mental health specialists and received 3-day training. Training and supervision were perceived as crucial to support health workers before and during the intervention. Technology served as a tool to structure the sessions, obtain and store patient data, present multi-media content, guarantee fidelity to the protocol and facilitate communication among members of the team. However, extra burden was mentioned by the health workers indicating the need of adjustments in their daily duties. Conclusions The PROACTIVE intervention was demonstrated to be feasible and accepted by both health workers and older adult participants. The qualitative assessments suggested improvements in training and supervision to ensure fidelity to protocol. To assess effectiveness a randomised controlled trial of the intervention will be conducted with the addition of improvements suggested by this qualitative study. Trial registration The pilot study of which the present study gives support to was registered at the Brazilian Clinical Trials, UTN code: U1111-1218-6717 on 26/09/2018.

Abstract Background The elderly population has been growing in most low- and middle-income countries (LMIC), and depression is a common condition among these populations. The lack of integration between mental health and primary healthcare services and the shortage of mental health specialists in the public health system contribute to underdiagnosis and undertreatment of depression. One of the strategies to reduce this gap is task shifting and collaborative care treatments. This study therefore aims to evaluate the effectiveness and cost-effectiveness of a collaborative care psychosocial intervention to improve the clinical management of depression among elderly people in poor neighbourhoods in Guarulhos, Brazil. Methods Two-arm, cluster randomised controlled trial with Basic Health Units as the clusters and a 1:1 allocation ratio. Twenty Basic Health Units have been randomly selected and randomised to control or intervention arms. We aim to recruit 1440 adults (72 per cluster) aged 60 years or over identified with depression (9-item Patient Health Questionnaire (PHQ-9) score ≥ 10). The control arm participants will receive an enhanced usual care, while the intervention arm participants will receive an enhanced usual care and a 17-week psychosocial intervention programme delivered at home by community health workers with the help of an application installed on tablet computers. The primary outcome is the proportion with depression recovery (PHQ-9

Abstract Background Six core behavioural risk factors (poor diet, physical activity, sedentary behaviour, alcohol misuse, smoking and unhealthy sleep patterns) have been identified as strong determinants of chronic disease, such as cardiovascular disease, diabetes and cancers. Smartphones have the potential to provide a real-time, pervasive, unobtrusive and cost-effective way to measure health behaviours and deliver instant feedback to users. Despite this, validity of using smartphones to measure these six key behaviours is largely unknown. The proposed systematic review aims to address this gap by identifying existing smartphone-based approaches to measure these health behaviours and critically appraising, comparing and summarizing the quality of their measurement properties. Methods A systematic search of the Ovid MEDLINE, Embase (Elsevier), Cochrane Library (Wiley), PsychINFO (EBSCOhost), CINAHL (EBSCOHost), Web of Science (Clarivate), SPORTDiscus (EBSCOhost) and IEEE Xplore Digital Library databases will be conducted from January 2007 to March 2020. Eligible studies will be those written in English that measure at least one of the six health behaviours of interest via a smartphone and report on at least one measurement property. The primary outcomes will be validity, reliability and/or responsiveness of these measurement approaches. A secondary outcome will be the feasibility (e.g. user burden, usability and cost) of identified approaches. No restrictions will be placed on the participant population or study design. Two reviewers will independently screen studies for eligibility, extract data and assess the risk of bias. The study methodological quality (or bias) will be appraised using an appropriate tool. Our results will be described in a narrative synthesis. If feasible, random effects meta-analysis will be conducted where appropriate. Discussion The results from this review will provide important information about the types of smartphone-based approaches currently available to measure the core behavioural risk factors for chronic disease and the quality of their measurement properties. It will allow recommendations on the most suitable and effective measures of these lifestyle behaviours using smartphones. Valid and reliable measurement of these behaviours and risk factor opens the door to targeted and real-time delivery of health behaviour interventions, providing unprecedented opportunities to offset the trajectory toward chronic disease. Systematic review registration PROSPERO: CRD42019122242

Background: People with cystic fibrosis (PWCF) have increased energy requirements. However, in recent years concerns have emerged regarding the ‘cystic fibrosis (CF) diet’ in terms of reliance on energy-dense, nutrient poor foods, which tend to be higher in saturated fat, sugar, and salt. These foods lack essential nutrients and are aetiologically linked with diet-related chronic diseases. The aim is to explore habitual dietary intakes in PWCF and (i) assess adherence to CF dietary guidelines and population specific healthy eating guidelines; (ii) derive a diet quality score and the inflammatory potential for the average diet consumed by PWCF and assess associations with patient reported outcome measures; (iii) assess drivers for current consumption patterns and enablers and barriers to eating a healthy diet. Methods: The aim is to recruit between 100-180 PWCF. A mixed methods study will be performed. Using three-day food diaries and food frequency questionnaires, aims (i) and (ii) will be addressed. The Dietary Approaches to Stop Hypertension (DASH) score and Healthy Eating Index-International (HEI-I) will derive diet quality scores. The Dietary Inflammatory Index (DII®) will ascertain inflammatory potential of the diet. Validated questionnaires will be used to report health related quality of life measures. Online focus groups and semi-structured interview with PWCF will address aim (iii). Conclusions: It is timely to revise dietary priorities and targets for CF. However, a greater understanding of what adults with CF currently consume and what they require in terms of nutrition and dietary guidance into the future is needed. In doing so, this research will help to clarify nutrition priorities and simplify the dietary aspects of CF treatment, thereby supporting adherence.

BackgroundPoor diet, alcohol use, and tobacco smoking have been identified as strong determinants of chronic diseases, such as cardiovascular disease, diabetes, and cancer. Smartphones have the potential to provide a real-time, pervasive, unobtrusive, and cost-effective way to measure these health behaviors and deliver instant feedback to users. Despite this, the validity of using smartphones to measure these behaviors is largely unknown. ObjectiveThe aim of our review is to identify existing smartphone-based approaches to measure these health behaviors and critically appraise the quality of their measurement properties. MethodsWe conducted a systematic search of the Ovid MEDLINE, Embase (Elsevier), Cochrane Library (Wiley), PsycINFO (EBSCOhost), CINAHL (EBSCOHost), Web of Science (Clarivate), SPORTDiscus (EBSCOhost), and IEEE Xplore Digital Library databases in March 2020. Articles that were written in English; reported measuring diet, alcohol use, or tobacco use via a smartphone; and reported on at least one measurement property (eg, validity, reliability, and responsiveness) were eligible. The methodological quality of the included studies was assessed using the Consensus-Based Standards for the Selection of Health Measurement Instruments Risk of Bias checklist. Outcomes were summarized in a narrative synthesis. This systematic review was registered with PROSPERO, identifier CRD42019122242. ResultsOf 12,261 records, 72 studies describing the measurement properties of smartphone-based approaches to measure diet (48/72, 67%), alcohol use (16/72, 22%), and tobacco use (8/72, 11%) were identified and included in this review. Across the health behaviors, 18 different measurement techniques were used in smartphones. The measurement properties most commonly examined were construct validity, measurement error, and criterion validity. The results varied by behavior and measurement approach, and the methodological quality of the studies varied widely. Most studies investigating the measurement of diet and alcohol received very good or adequate methodological quality ratings, that is, 73% (35/48) and 69% (11/16), respectively, whereas only 13% (1/8) investigating the measurement of tobacco use received a very good or adequate rating. ConclusionsThis review is the first to provide evidence regarding the different types of smartphone-based approaches currently used to measure key behavioral risk factors for chronic diseases (diet, alcohol use, and tobacco use) and the quality of their measurement properties. A total of 19 measurement techniques were identified, most of which assessed dietary behaviors (48/72, 67%). Some evidence exists to support the reliability and validity of using smartphones to assess these behaviors; however, the results varied by behavior and measurement approach. The methodological quality of the included studies also varied. Overall, more high-quality studies validating smartphone-based approaches against criterion measures are needed. Further research investigating the use of smartphones to assess alcohol and tobacco use and objective measurement approaches is also needed. International Registered Report Identifier (IRRID)RR2-10.1186/s13643-020-01375-w

BackgroundMental disorders are normally diagnosed exclusively on the basis of symptoms, which are identified from patients’ interviews and self-reported experiences. To make mental health diagnoses and monitoring more objective, different solutions have been proposed such as digital phenotyping of mental health (DPMH), which can expand the ability to identify and monitor health conditions based on the interactions of people with digital technologies. ObjectiveThis article aims to identify and characterize the sensing applications and public data sets for DPMH from a technical perspective. MethodsWe performed a systematic review of scientific literature and data sets. We searched 8 digital libraries and 20 data set repositories to find results that met the selection criteria. We conducted a data extraction process from the selected articles and data sets. For this purpose, a form was designed to extract relevant information, thus enabling us to answer the research questions and identify open issues and research trends. ResultsA total of 31 sensing apps and 8 data sets were identified and reviewed. Sensing apps explore different context data sources (eg, positioning, inertial, ambient) to support DPMH studies. These apps are designed to analyze and process collected data to classify (n=11) and predict (n=6) mental states/disorders, and also to investigate existing correlations between context data and mental states/disorders (n=6). Moreover, general-purpose sensing apps are developed to focus only on contextual data collection (n=9). The reviewed data sets contain context data that model different aspects of human behavior, such as sociability, mood, physical activity, sleep, with some also being multimodal. ConclusionsThis systematic review provides in-depth analysis regarding solutions for DPMH. Results show growth in proposals for DPMH sensing apps in recent years, as opposed to a scarcity of public data sets. The review shows that there are features that can be measured on smart devices that can act as proxies for mental status and well-being; however, it should be noted that the combined evidence for high-quality features for mental states remains limited. DPMH presents a great perspective for future research, mainly to reach the needed maturity for applications in clinical settings.

Abstract Background Depression is a common and recurrent condition among older adults and is associated with poor quality of life and increased health care utilization and costs. The purpose of this pilot study was to assess the feasibility of delivering a psychosocial intervention targeting depression, and to develop the procedures to conduct a cluster randomized controlled trial among older adults registered with primary care clinics in poor neighbourhoods of São Paulo, Brazil. Methods We conducted a pilot study of a two-arm cluster, non-randomized controlled trial. Two primary care clinics adhering to the Family Health Strategy were allocated to either the intervention or the control arm. In the control arm, patients received enhanced usual care consisting of staff training for improved recognition and management of depression. In the intervention arm, alongside the enhanced usual care, patients received a 17-week psychosocial intervention delivered by health workers assisted with an application installed in a tablet. Results We randomly selected 579 of 2020 older adults registered in the intervention clinic to participate in the study. Among these individuals, 353 were assessed for depression and 40 (11.0%) scored at least 10 on the PHQ-9 and were therefore invited to participate. The consent rate was 33/40 (82%) with a resulting yield of 33/579 (5.7%). In the control arm, we randomly selected 320 older adults among 1482 registered in the clinic, 223 were assessed for depression and 28 (12.6%) scored 10 or above on the PHQ-9. The consent rate was 25/28 (89%), with a resulting yield of 25/320 (7.8%). Of the 33 who consented in the intervention arm, 19 (59.4%) completed all sessions. The mean PHQ-9 at follow-up (approximately 30 weeks after inclusion) were 12.3 (SD = 3.7) and 3.8 (SD = 3.9) in the control and intervention arms, respectively. Follow-up rates were 92 and 94% in control and intervention arms, respectively. Conclusions Identification and engagement of clinics, randomization, recruitment of individuals, measures, and baseline and follow-up assessments all proved to be feasible in primary care clinics in São Paulo, Brazil. Results support the development of a definitive cluster randomized controlled trial. Trial registration This study was retrospectively registered with Registro Brasileiro de Ensaios Clínicos (ReBEC), number RBR-5nf6wd. Registered 06 August 2018.

In this paper we introduce a new Android library, called ULTEMAT, for the delivery of ecological momentary assessments (EMAs) on mobile devices and we present its use in the MoodBuster app developed in the H2020 E-COMPARED project. We discuss context-aware, or event-based, triggers for the presentation of EMAs and discuss the potential they have to improve the effectiveness of mobile provision of mental health interventions as they allow for the delivery of assessments to the patients when and where these are most appropriate. Following this, we present the abilities of ULTEMAT to use such context-aware triggers to schedule EMAs and we discuss how a similar approach can be used for Ecological Momentary Interventions (EMIs).

An amendment to this paper has been published and can be accessed via the original article.

In this paper, the authors investigate the role that smart devices, including smartphones and smartwatches, can play in identifying activities of daily living. A feasibility study involving N = 10 participants was carried out to evaluate the devices’ ability to differentiate between nine everyday activities. The activities examined include walking, running, cycling, standing, sitting, elevator ascents, elevator descents, stair ascents and stair descents. The authors also evaluated the ability of these devices to differentiate indoors from outdoors, with the aim of enhancing contextual awareness. Data from this study was used to train and test five well known machine learning algorithms: C4.5, CART, Naïve Bayes, Multi-Layer Perceptrons and finally Support Vector Machines. Both single and multi-sensor approaches were examined to better understand the role each sensor in the device can play in unobtrusive activity recognition. The authors found overall results to be promising, with some models correctly classifying up to 100% of all instances.

Current mobile devices allow the execution of sophisticated applications with the capacity for identifying the user situation, which can be helpful in treatments of mental disorders. In this paper, we present SituMan, a solution that provides situation awareness to MoodBuster, an ecological momentary assessment and intervention mobile application used to request self-assessments from patients in depression treatments. SituMan has a fuzzy inference engine to identify patient situations using context data gathered from the sensors embedded in mobile devices. Situations are specified jointly by the patient and mental health professional, and they can represent the patient’s daily routine (e.g., “studying”, “at work”, “working out”). MoodBuster requests mental status self-assessments from patients at adequate moments using situation awareness. In addition, SituMan saves and displays patient situations in a summary, delivering them for consultation by mental health professionals. A first experimental evaluation was performed to assess the user satisfaction with the approaches to define and identify situations. This experiment showed that SituMan was well evaluated in both criteria. A second experiment was performed to assess the accuracy of the fuzzy engine to infer situations. Results from the second experiment showed that the fuzzy inference engine has a good accuracy to identify situations.